Although efforts to improve the quality of diabetes care have increased over the past decade, quality remains suboptimal, especially for minority patients. We propose to test the effectiveness of an intervention designed to increase patient participation in treatment decisions on diabetes quality, particularly HbA1c values, among Mexican American, Vietnamese American, and non-Hispanic white patients in community-based outpatient clinics. This intervention, now called "Coached Care," has been well tested in multiple randomized controlled trials among less diverse patient groups, including those with diabetes, hypertension, breast cancer, and rheumatoid arthritis. Different from self-management or more traditional patient education, this intervention focuses on: 1) providing patients with individualized treatment information based on an algorithm for diabetes management coupled with a patients' medical records; 2) skills for using the information to negotiate a treatment regimen more consistent with the patients' preferences and lifestyle; and 3) preparation of patients immediately preceding the office visit for a more active role in care. In order to adapt this intervention to the unique needs of Mexican and Vietnamese American patients, we propose to have volunteer "coaches," drawn from the patient's community who themselves have diabetes. These coaches will prepare patients prior to office visits to participate in treatment decisions and identity barriers to effective participation. The coaches will reinforce negotiation skills between office visits. They will also help patients and physicians to identify and resolve culture-specific conflicts with effective treatment implementation and disease management. In a large randomized controlled trial, we will test the impact of the intervention on clinical outcomes and on the quality of interpersonal for each of the three patient groups. We will measure the costs of the intervention relative to any observed changes in the quality of diabetes care to assess the feasibility of widespread implementation of the intervention. If the intervention is effective and feasible, it could serve as model for improving chronic disease care in minority communities in settings with limited resources. [unreadable] [unreadable]